Antiretroviral (ARV)-based HIV-1 prevention - including 1) ARV treatment (ART) to reduce the infectiousness of HIV-1 infected persons and 2) oral and topical pre-exposure prophylaxis (PrEP) for uninfected persons to prevent HIV-1 acquisition - is one of the most promising approaches for dramatically decreasing HIV-1 spread. Critical unanswered questions for successful delivery of ARV-based HIV-1 prevention include how to target ART and PrEP to realize maximum population HIV-1 prevention benefits, and for those targeted, whether HIV-1 infected persons at earlier stages of infection would accept ART to reduce their risk for transmitting HIV-1, highest-risk HIV-1 negative persons would use PrEP, and both would sustain high adherence needed to achieve high effectiveness. Stable, heterosexual, African HIV-1 serodiscordant couples (i.e., in which one member is HIV-1 infected and the other uninfected) face high risk of HIV-1 transmission and will be a prime target for ARV-based prevention. Since 2004, we have recruited >8500 African HIV-1 serodiscordant couples into prospective HIV-1 prevention studies, including 4700 couples from Kenya and Uganda into the Partners PrEP Study, an ongoing trial of oral tenofovir-based PrEP. HIV-1 serodiscordant couples will be a primary target globally for ARV-based prevention interventions, with the opportunity to use ARVs as treatment or as PrEP for prevention. Successful implementation of ARV-based HIV-1 prevention among HIV-1 serodiscordant couples will need to target delivery to highest risk couples, define optimal strategies for targeting for ART for HIV-1 infected partners versus PrEP for HIV-1 uninfected partners, respond to couples' preferences for prevention interventions, especially preferences for ART versus PrEP, and achieve high uptake and sustained adherence. We propose a series of integrated studies to optimize targeted delivery and sustained use of ARV-based HIV-1 prevention in African HIV-1 serodiscordant couples. We will analyze data from our previous cohorts of HIV-1 serodiscordant couples, implement additional data collection in our ongoing Partners PrEP Study, and test our hypotheses in a new cohort of highest-risk couples from Kenya and Uganda. We have assembled a multinational and multidisciplinary project team has the expertise and experience to complete the proposed work. We will: 1) develop a prediction risk score to identify priority HIV-1 serodiscordant couples to target for ARV-based HIV-1 prevention, based on highest risk for HIV-1 seroconversion, 2) use mathematical modeling to define optimal targeting of early ART of HIV-1 infected partners versus PrEP for HIV-1 uninfected partners for prevention in couples, 3) assess user preferences among high-risk HIV-1 serodiscordant couples about earlier ART initiation for HIV-1 infected partners versus PrEP for HIV-1 uninfected partners, and 4) prospectively ascertain initiation of and adherence to, ARV-based HIV-1 prevention strategies, including facilitators and barriers, among high-risk HIV- 1 serodiscordant couples. The proposed project offers an unprecedented opportunity to explore the critical public health questions surrounding ARV-based HIV-1 prevention in a target population for implementation. The project is novel and significant in its integration of biobehavioral and sociocultural research, focus on primary HIV-1 prevention and prevention of transmission from persons with HIV-1, target of dyads to identify appropriate prevention approaches, and aim to identify barriers and facilitators for dissemination of effective HIV-1 prevention.